ing this approach. Blood collected in the drains within the first 2–4 postopera- tive hours can also be processed and reinfused with the cell saver system. Pharmacological Measures Tranexamic acid or aprotinin [81] used with the induction of anesthesia has been reported both in adults and children to reduce blood losses in spinal procedures. Because of its price (1 g tranexamic acid costs C$19.35 vs. C$210 per allogenic blood unit vs. C$338 per autologous blood unit vs. C$344.40 per vial of 500000 U of aprotinin), good tolerance and effectiveness, we and others [54, 65] prefer tra- nexamic acid in a protocol of 15–50 mg/kg in a bolus with the induction of anes- thesia plus an infusion of 1 g/h or boluses of 10–25 mg/kg every 3 h intraoperati- vely and then q8 h for the first 24 h postoperatively. An increase in coagulability, changes in kaolin/Celite times or severe allergic reactions associated with the use of aprotinin have not been reported with tranexamic acid [26]. Recently, the use of aprotinin was associated with a doubling of the risk of renal failure, a 55% increased risk of myocardial infarction and a 181% increase in the risk of stroke in cardiac surgery when compared to tranexamic acid [45]. Desmopressin has not proven useful in decreasing blood losses [76] in idiopathic scoliosis surgery. Anemia/hemodilution and low CHA increase the risk of ION We do not use hemodilution since there is no demonstrated advantage of add- ing it to patients having CHA and antifibrinolytics. More importantly, ION seems to be much more likely to occur when combining anemia (or hemodilution) and low CHA. Blood Transfusion and Coagulation Factor Substitution The question of when to start transfusing blood products in spine surgery boils down to what are the thresholds for the red cells (RBCs), platelets, plasma and factors. Blood is separated in blood banks into its components to optimize the use of resources by allowing blood subproducts to be transfused in different patients. Two different approaches to blood component replacement have been recommended. The first is to transfuse fresh frozen plasma (FFP) and platelets prophylacticallyafteracertainnumberofunitsofRBCshavebeentransfused. However, there is no agreement on the optimal ratios; these vary widely, ranging from 1:10 to 2:3 for FFP:RBCs and from 6:10 to 12:10 for platelets:RBCs. The second approach is to transfuse FFP, platelets or cryoprecipitate only when there is clinical or laboratory evidence of coagulopathy; for instance, when there is microvascular bleeding, a prothrombin time (PT) or a partial thromboplastin time (PTT) >1.5 times the normal value, thrombocytopenia with a platelet count <50000–100000/l or a fibrinogen concentration <100 mg%. The following are recommendations from international publications summa- rized by Leal-Noval [42] and the American Society of Anesthesiologist Task Force on Perioperative Blood Transfusions 2005 (www.asahq.org). RBC Concentrates Transfusion Criteria Hb <8 g% Note: 10 ml/kg of RBC concentrate will increase the Hb by 1–2 g% or 3–6 points of hematocrit Hb between 8 and 10 g% in normovolemic patients, but with clinical signs of myocardial, cerebral, or respiratory dysfunction; and intraoperative hemorrhage, i.e., bleeding of 10 ml/kg in the first hour or 5ml/kg×hinthefirst3h(averaged) Intraoperative Anesthesia Management Chapter 15 403 FFP Transfusion Criteria Patients with active bleeding and: Each unit of FFP contains 2–4 mg of fibrinogen/ml; therefore each FFP unit delivers the equivalent of 2 units of cryoprecipitate PT or PTT 1.5 times that of control subjects; International Normalized Ratio (INR) >2.0 massive transfusion of RBC concentrates >30 ml/kg of packed red cells previous treatment with coumadin derivatives and unscheduled surgery (to give FFP 5–8 ml/kg) correction of factor deficiencies when specific factors are unavailable (to give FFP 10–15 ml/kg) heparin resistance (antithrombin III deficit) Platelet Transfusion Criteria Patients with severe hemorrhaging and: diffuse bleeding suggestive of platelet dysfunction platelet count <50000–100000/l massive transfusion of RBC concentrates normal platelet count and platelet dysfunction (antiplatelet agents, throm- basthenia, uremia, etc.) Cryoprecipitate and Factor Transfusion Criteria patients with active bleeding and fibrinogen <80 mg% bleeding patients with von Willebrand’s disease in absence of specific con- centrates Note: Each unit of cryoprecipitate contains 150–250 mg of fibrinogen. The start- ing dose is 1 unit for 10 kg body weight to increase fibrinogen level by 50 mg% (the hemostatic level is around 100 mg%). Cryoprecipitate does not contain Fac- tor V. Therefore, it should not be the sole replacement therapy for disseminated intravascular coagulopathy (DIC), which is almost always associated with a vari- ety of factor deficiencies and thrombocytopenia. Intermediate purity Factor VIII concentrates are preferred for von Willebrand’s disease and recombinant or highly purified Factor VIII concentrate for hemophilia A because of its greater efficacy and safety. The intermediate purity concentrate contains significant therapeutic quantities of the von Willebrand’s component of Factor VIII, whereas the high purity preparations contain principally the hemophilia A com- ponent of Factor VIII. Transfusion Criteria for rFVIIa rFVIIa is approved in many countries for patients with hemophilia and inhibitors (antibodies) to coagulation factors VIII or IX. High circulating concentrations of FVIIa, achieved by exogenous administration, initiate hemostasis by combining with tissue factor at the site of injury, producing thrombin, activating platelets and coagulation factors II, IX and X, thus providing for the full thrombin burst that is essential for hemostasis. This “bypass” therapy has led some clinicians to use rFVIIa “off-label” for disorders of hemostasis other than hemophilia. The Israeli Multidisciplinary rFVIIa Task Force published their guidelines for its use in uncontrolled bleeding [47], which recommended that optimal conditions (fibrinogen concentration >50 mg%, platelet count >50000/l, pH >7.2) should be achieved before the administration of rFVIIa. There are no clear recom- mended doses yet for rFVIIa. A wide range of between 50 and 200 μg/kg has been 404 Section Peri- and Postoperative Management advocated. Because of its clearance (35 ml/kg/h), it is suggested to repeat the dose every 2 h in case of persistent hemorrhage [82]. Massive transfusion canbedefinedastheacutereplacementofmorethanone bloodvolumewithin6h.Inpreviouslyhealthyadults,coagulationdefects develop primarily from dilution of protein coagulation factors and platelets when crystalloid, colloid and RBCs are used to replace lost volume. Coagulopa- thy associated with massive transfusion is clinically characterized by the pres- ence of microvascular bleeding or oozing from the mucosae, wound and punc- ture sites. The development of acidosis, DIC, hypothermia and, rarely, a hemo- Massive transfusions may result in acidosis, DIC, hypothermia and hemolytic transfusion reactions lytic transfusion reaction may accompany massive transfusion and complicate the ability to diagnose the coagulopathy. While thrombocytopenia may develop in massively transfused patients, administration of platelets should be reserved for the patient exhibiting microvascular bleeding and a platelet count of less than 50000/l. In the massively transfused patient, clinical bleeding associated with coagulation factor deficiencies is unlikely until factor activity levels fall below 20%ofnormal.Thisusuallydoesnotoccuruntilgreaterthanonebloodvolume has been replaced. FFP may be administered for correction of microvascular bleeding in patients transfused with more than one blood volume. PT and PTT along with platelet count and fibrinogen level should guide the use of component therapy. Whole blood clotting analysis, as seen with thromboelastography, pro- vides a dynamic picture of the entire clotting process. Some potential metabolic problems resulting from blood transfusion are hyperkalemia, hypocalcemia, cit- rate toxicity, hypomagnesemia, acidosis and impaired oxygen-carrying capacity of hemoglobin. The electrocardiogram should be monitored in all patients for signs of electrolyte abnormality during rapid infusions. Hyperkalemia exacer- bates the cardiovascular effects of hypocalcemia. Administration of calcium rap- idly antagonizes hyperkalemia by promoting transfer of potassium intothe cells. Intraoperative Spinal Cord Monitoring Patients undergoing corrective surgery for deformity are at a higher risk of spinal cord injury. Similarly, patients who have sustained an incomplete traumatic spi- nal cord injury are at risk of further damage. Neurological deterioration can occur because of ischemia of the neural structures secondary to mechanical com- pression and/or vascular stretching. Monitoring must be performed by an expe- Spinal cord monitoring requires clinical practice for its effective use rienced team and the surgeon must be interested in acting on the findings [18]. Teamwork and communication between the electrophysiology technician, anes- thesiologist and surgeon are necessary to make spinal cord monitoring useful for the patient. Important facts regarding anesthesia stability and depth, hemody- namics, blood volume,bloodflowautoregulation ofthespinalcord and tempera- ture must be considered. MAP below 60 mmHg or hypovolemia can result in sig- nificant changes in SSEPs [55, 57]. During surgery, a MAP of 60–65 mmHg is usually maintained to reduce blood loss. Drops in temperature can affect SSEP waveforms [46]. If the limbs, brain or spinal cord become cooler during surgery, SSEP latencies will increase without an actual injury to the neural pathway. The anesthesia goals to facilitate neuromonitoring are highlighted in Table 4.Anelec- Table 4. Goals of anesthesia management to facilitate neuromonitoring tight and stable hypotensive blood pressure control normal end tidal CO 2 normothermia normovolemia stable depth of anesthesia compatible with neuromonitoring Hb level above 7 g % Intraoperative Anesthesia Management Chapter 15 405 tric line interference of 60 Hz coming from the operating room table or other electric equipment may severely affect the SSEP recordings [56]. In the presence of intraoperative spinal cord monitoring (IOM), neurological deficits after spine surgery relate to [56]: type of procedure the surgeon’s experience of spine surgery the surgeon’s experience using SSEP the technician’s experience (experience with less than 100 cases doubled the deficits compared with >300 cases) Low (or narrower) cut filtering (30 Hz to 1 kHz) is better than 1 Hz to 5 kHz). Anesthetic Effects on SSEPs Intravenous opiates may increase the latency of SSEPs Halogenated anesthetics produce a dose-related reduction in amplitude and an increase in the latency of responses to SSEPs [69]. Nitrous oxide adds more intense changes in cortical SSEP recording than those of halogenated drugs and in fact they are synergic with isoflurane when used together. Sevoflurane, desflu- rane or mixtures of N 2 O opiates may be used during SSEP monitoring as long as the concentration of the inhaledagentsiskeptlow(below0.7 MAC) and stable to avoid artificial effects due to changes in depth of anesthesia. Subcortical record- ings (from C2) are relatively resistant to the depressing effects seen when cortical level recordings are made. Cortical evoked potential (CEP) changes related to deepening anesthesia may be indistinguishable from spinal cord injury. For this reason, subcortically generated SSEP recordings should be obtained to corrobo- rate CEP changes, while peripheral nerve responses should be recorded to ensure that the adequacy of stimulation has not changed to account for the CEP change. Intravenous opiates used with inhaled agents in clinical anesthesia produce little impact on the amplitude of EEG; however,they may increase the latency of SSEPs. This small effect of the systemic opiates on latency recordings seems to be μ- receptor dependent and occurs at a supraspinal level since spinal/epidurally administered morphine or fentanyl minimally affects SSEPs [67]. Ketamine is an NMDA antagonist, which has become more popular lately as part of a multi- modal anesthetic approach. Ketamine is known to increase amplitude responses in cortical SSEPs as well as spinal and muscle recordings after spinal activation [39]. Nonetheless, ketamine could be a problem when a WUT is required. A simi- lar observation about SSEPs has been made with etomidate [37]. Thiopental is a barbiturate and poses no problems for monitoring neurological parameters dur- ing spine surgery after the rapid redistribution of the single induction dose. Short-acting benzodiazepines are combined with opiates or ketamine as part of a balanced technique. Induction and maintenance of anesthesia with midazolam induces negligible changes to cortical SSEP recordings [66]. Combinations of midazolam-fentanyl and midazolam-ketamine along with N 2 O have been found equally appropriate in spine surgery and SSEP recording [39]. Propofol is dependable for both the induction and maintenance of anesthesia with a very predictable pharmacodynamic response when used with target controlled infu- sions (TCIs). Propofol slightly depresses the amplitude of SSEPs at the brain cor- tex level with negligible action at clinical doses on spinal cord physiology. Propo- fol is regarded as a very good alternative for anesthesia during functional moni- toring in spine surgery [69]. Muscle relaxants do not affect SSEPs and in fact they might enhance the SSEP signal by decreasing electric noise by eliminating mus- cle artifacts. Epidural/intrathecal, but not i.v., local anesthetics increase SSEP latency and are contraindicated because of their direct effect in spinal cord con- duction [36]. 406 Section Peri- and Postoperative Management Red Flags in SSEP Recordings SSEP recordings can be affected in two dimensions: amplitude and/or latency. A 50% decrease in amplitude and/or a 10% or 2-ms increase in latency in a hemo- dynamically stable, normothermic patient are considered as indicators of spinal cord insult [56]. In this case, counteractive measures encompass surgical and anesthetic reactions (see Table 5).Changesinrecordingsthatdonotreverseto normal after corrective measures and are still present at the end of the procedure correlate with new postoperative nerve deficits [72]. Table 5. Course of action suggested for deteriorating neuromonitoring Surgical interventions Anesthetic interventions reduction of correction increase in blood pressure removal of implant correction of anemia correction of hypovolemia normalization of temperature lighter anesthesia level IV steroids normalization of CO 2 Anesthetic Effects on MEPs MEPs are obtained by transcranial electrical (tcEMEP) or magnetic (tcMMEP) stimulation of the motor cortex and recordings are made in muscles or periph- eral nerves. Stimulation can also be made at a high epidural level next to the spi- nal cord. In patients with spinal cord deficits, MEPs can be present when SSEPs are absent and vice versa. Repetitive transcranial stimulation (trains of three to five impulses as opposed to a single stimulus) can overcome some of the depres- sant actions of anesthetics by temporal summation of the descending input on the motoneurons. MEP changes during spine surgery correlate well with neuro- logical outcome. MEPs are complementary to SSEPs in reducing spinal cord risk MEP changes predict neurological outcome of damage in complex spinal surgery. tcMMEP seems to be more affected by anesthetics than tcEMEP [69]. MEPs may allow adequate recordings of patients who are otherwise “unmonitorable” by SSEPs. MEP signals should have an amplitudeofatleast50μVbeforetheyareconsideredtobe“monitorable.”Keta- mine based anesthesia allows for appropriate MEP recording because of its mini- mal depressing actions. Barbiturates must be avoided if early recording of tcMMEPs is required because up to 45 min of deep depression has been reported [21]. Midazolam and thiopental share the same depressing effect on tcMMEPs, so these agents are not recommended when that kind of monitoring is to be used [31]. Complete motor blockade will prevent muscle response and recording of cranial or spinal cord induced MEPs. Partial neuromuscular blockade with con- tinuous and stable infusions of muscle relaxants to keep a train-of-four of 3/4 has been successfully reported [38]. Constant evaluation with nerve stimulators or closed-loop systems might produce a level of relaxation compatible with optimal recording of MEPs and very good surgical conditions. These evoked potentials are large responses clear of signal averaging that can provide the surgeon with good feedback. MEPs may be contaminated by sudden patient movement and anesthetic agents. Intraoperative Anesthesia Management Chapter 15 407 Red Flags in MEP Recordings A rapid and permanent decrease in signal amplitude larger than 50%, or a 100 V increase in the threshold of the MEP muscle response, is indicative of a neural compromise [50, 84] with potential neurological consequences. Nerve Root Monitoring SSEPs and MEPs are less likely to alert the surgeon about single root potential damage than techniques monitoring that particular root. Electrical stimulation of screws placed in the pedicles can confirm correct placement or signal a breach in the bone cortex by lowering the current needed to activate a sustained neuro- tonic electromyogram (EMG) discharge from the muscles innervated by that root [13]. Some consider there is a malpositioned screw when a recording of com- pound muscle action potential is obtained of less than 10 mA and 200 μs pulse width stimulation. No response with intensities above 15 mA was found to be 98% accurate for properly implanted screws [20]. The reported rate for false neg- atives and sensitivity is 8% and 93%, respectively [44, 83]. This technique also allows for continuous EMG recording, so that changes can be observed on decompressing the roots, cage positioning and rod placement. No neuromuscu- lar relaxant drug (NMB) effects have to be observed (at least three out of four twitches in the train-of-four) over the period of surgical EMG monitoring. Wake-up Test A WUT consists of stopping the anesthetics after surgical spine manipulation to assess the motor function of the spinal cord and nerve roots. Usually the spinal cord, brachial plexus roots, and L5 and S1 can be evaluated by asking the patient to move their hands and feet. The WUT is an outstanding procedure for ascer- taining corticospinal and motoneuron integrity. In experienced hands a WUT is quick, reliable, safe and reproducible. It requires 5–15 min notice from the sur- geon to conduct it. Many spine surgeons feel comfortable omitting a WUT when reliable data with SSEPs and MEPs are obtained and maintained. The WUT is currently performed when there is no SSEP/MEP data available or in circum- stances where these methods are not reliable. The limitations of the WUT are: Spinal cord monitoring has replaced WUTs in many centers intermittent rather than continuous monitoring not applicable in mentally handicapped patients not feasible in small children preexisting severe spinal cord damage (incomplete lesion) Venous embolism, corneal damage, loss of vascular access, violent wake-up, acci- dental extubation or hardware dislodgement is unlikely when the test is con- ducted in skilled hands. The WUT technique requires training and practice to master and be used with confidence. A normal WUT with posterior column damage or “false negative” (with documented intraoperative SSEP deficit) has been reported by Ben-David [6]. This is not a true false negative because SSEPs and the classic WUT are aimed at different anatomic structures: dorsal column and anterior spinal cord blood supply. We have refined a WUT that allows us to test both sensory/proprioceptive and motor components in a reliable and quiet fashion. 408 Section Peri- and Postoperative Management End of Anesthesia Planning for postoperative pain control, elective postoperative ventilatory sup- port and postoperative destination should be conducted before starting the sur- gery. However, emergency cases and unexpected intraoperative events might require fast intraoperative decision-making. Ideally patients should be quickly regaining the ability to follow commands to assess their neurological status, be comfortable with coughing to clear secretions and starting with physiotherapy. The provision for pain management is discussed in the next section. Elective and lastminutedecisionstokeepthepatientintheintensivecareunitareshownin Table 6. Patients with major comorbidities before surgery and/or unexpected adverse intraoperative events account for most indications for the postoperative ICU. Which patients should have postoperative ventilation? Most spine surgery patientsareextubatedonthetableattheendofthesurgery.Considerationfor postoperative mechanical ventilation should be given to patients undergoing neuromuscular scoliosis correction, with preoperative respiratory or cardiac The need for postoperative mechanical ventilation must be considered prior to surgery dysfunction, having intraoperative hemodynamic and respiratory instability, with unexpected decreases in body temperature, with difficult airway access, or with slow recovery from anesthesia [60]. Although it is not our regular practice, some groups suggest elective ventilation for a few hours after C-spine surgery to make certain no airway compromise by hematoma is present after surgery and before extubation. Table 6. Perioperative considerations regarding overnight ICU requirement Preoperative reasons Intraoperative reasons Postoperative reasons preoperative severe respiratory impairment cervical spine surgery: laryngeal nerve damage or hematoma respiratory failure mental disability hemodynamic instability hemodynamic instability congestive heart failure continued correction of hypovolemia special monitoring requirements chronic obstructive pulmonary disease surgical complications chronic renal failure coagulopathy muscular dystrophy anesthetic complications patient coming from ICU hypothermia Postoperative Pain Management Postoperative pain and gastrointestinal dysfunction (nausea, vomiting, ileus, con- stipation and anorexia) secondary to analgesics and other drugs are among the main factors delaying the recovery process in spinal surgery. The goals of postop- erative pain control therapies are to enhance recovery and decrease complica- tions rather than just to decrease pain measured scores.Challengesrelatetopre- operative pain and opioid tolerance, cognitive impairment, extremes of life and difficulties assessing the symptoms and the results of the treatments applied. A multimodal approach is recommended, involving acetaminophen, low-dose NSAIDs, systemic opioids, wound infiltration with local anesthetics and coadju- vants (i.e., low-dose ketamine, stool softeners and gabapentin). The requirements of preoperative opioids do not disappear right after the surgery. It might take weeks. Therefore, it is recommended to restart them as baseline analgesia as soon as thepatient can receive them orally or to replace them temporarily intravenously. Multimodal Analgesia. Acetaminophen is extremely well tolerated and can be used before beginning the surgery per rectum, per os or intravenously (as propa- Intraoperative Anesthesia Management Chapter 15 409 racetamol) in doses of 15 mg/kg every 4–6 h. Metamizol (Dypirone) is an excel- lent alternative to acetaminophen at the same dose regimen provided the patient The postoperative use of NSAIDs remains amatterofdebate is not allergic to it and has no bone marrow disease. The postoperative use of NSAIDs has been the subject of heated controversy in the literature because of data coming from animal studies and retrospective human chart reviews. There is not a single prospective randomized trial on spine surgery in humans demon- strating a higher incidence of malunion or a slower consolidation secondary to shortuse(3–5days)ofNSAIDs.Onthecontrary,theliteratureshowssimilar surgical outcomes with better pain control in patients who received ketorolac at less than 110 mg/day after spine procedures [22, 51, 61]. These analyses have actually emphasized that preoperative smoking increases the risk of malunion by 8–15 times. NSAIDs only become an issue when they are used in high doses in smokers. If the patient is going to have low molecular weight heparin postsurgery (uncommon in spine procedures), it seems safer to use a COX-2 specific such as celecoxib (rule out cardiovascular contraindications). Wound infiltration at the beginning and the end of the operation greatly reduces the amount of anesthetics and opioids required in the first few hours after surgery, allowing patients to be scheduled to go home the same day (i.e., after disc surgery) and a smoother tran- sition and discharge. P atient controlled analgesia (PCA), nurse or parent assisted PCA or regular subcutaneous opioids are the most commonly used anal- gesia technique after spine procedures. Side effects are often prominent includ- ing gastrointestinal, excessive sedation, respiratory depression and poor inci- dental pain relief. The advantages of using epidural analgesia after scoliosis surgery ( Fig. 5)have been reported by Blumenthal [9] and Tobias [78]. Both methods (PCA and epidu- ral) provided efficient postoperative analgesia. However, the double epidural catheter technique provides better postoperative analgesia, earlier recovery of bowel function, fewer side effects, and higher patient satisfaction. Figure 5. Cerv icothoracic epidural catheter Epidural catheter at the level of C7/T1 allows for excellent pain control in cases with posterior fusion and/or a transthoracic approach. 410 Section Peri- and Postoperative Management Recapitulation Communication. Anesthesiologists with special ex- pertise in spine surgery play an important role in the perioperative team in charge of patients. The anesthesiologist will lay out a plan to manage anes- thesia in each case, but this plan must be closely in- tegrated into the surgical plan. Therefore, the anes- thesiologist must be involved before surgery to permit a team plan for the case, no matter how sim- ple it may seem. Goals in spinal surgery. Critical aspects of the intra- operative anesthesia care are airway management, positioning on the operative table, techniques to minimize surgical bleeding, pain control and organ perfusion. Techniques to control bleeding must be balanced against ocular complications and cord function and perfusion. Techniques to secure the airway must be balanced against spinal cord injury. Techniques to achieve proper pain control postsur- gery must be balanced against effective bone fu- sion and clean healing. Induction of anesthesia. In this period, the critical issues are airway control and hemodynamic stabili- ty. Patients with an unstable cervical spine require careful fiberoptic tube placement, avoiding drops in blood pressure that might further jeopardize the cord condition. Patients coming for transthoracic surgical approaches might require lung deflation by using a bronchial blocker or other device to facil- itatesurgicalexposure.Thereisnoevidencetosup- port the use of armored endotracheal tubes. Antibi- otic prophylaxis before starting the operation is mandatory in most spine surgery cases to preclude colonization of implants. Maintenance of anesthesia. In the maintenance period of major spine cases, controlled hypoten- sion to MAP not below 60 –65 mm Hg along with tranexamic acid is an efficacious means to control bleeding and allow for a drier surgical field. Intrao- perative neuromonitoring requires stable tempera- ture, anesthesia depth with low doses of gases or TIVA and good cord perfusion. Guidelines are pro- vided for transfusions in the spine surgery scenario as well as a clear and simple description of the wake-up test for places without an SSEP machine. In simple cases of day surgery procedures, the goals are rapid recovery of anesthesia without nausea, vomiting and pain. Local anesthesia infiltration before the surgery and at the end facilitates an an- esthetic approach with minimal opioids. End of anesthesia. At the conclusion of the anes- thesia and surgery, the issues are pain control and again airway management. Multimodal analgesia along with epidural catheters offers excellent re- sults with low morbidity and high levels of patient (and surgeon) satisfaction. NSAIDs in low doses (ke- torolac <90 mg/day or celecoxib <200 mg/day) and for less than 72 h postoperatively are a safe and effective part of the cocktail as long as the patient is a nonsmoker. The decision to keep the patient intu- bated in the first few hours after C-spine or major spine operations should rely on the clinical assess- ment by the team regarding the physiologic and anatomic conditions of the individual patient. Key Articles Lauer KK (2004) Visual loss after spine surgery. J Neurosurg Anesthesiol 16:77 – 79 Brief review of the topic with excellent and concise information to understand why this complication occurs in spine surgery. Sessler D (2001) Complications and treatment of mild hypothermia. Anesthesiology 95:531 – 43 The author analyzes the clinical implications of perioperative hypothermia. An impor- tant paper that presents very practical information about the deleterious effects of mild hypothermia on infectious, metabolic and hemostatic aspects usually unknown to many clinicians. Tobias JD (2004) Strategies for minimizing blood loss in orthopedic surgery. Semin Hematol 41(1):145 – 56 Comprehensive review of the current techniques to preserve blood in spine surgery. Intraoperative Anesthesia Management Chapter 15 411 Key Articles www.asahq.org/publica tionsAndServices/transfusion.pdf This web site of the American Society of Anesthesiology presents very well documented guidelines about blood product therapy in the perioperative period. It is frequently updated with new information and is easy to read. Duffy G, Neal KR (1996) Differ ences in postoperative infection rates between patients receiving autologous and allogeneic blood transfusions: a meta-analysis of published randomized and n onrandomized studies. Transfus M ed 6(4):325 – 28 The authors reviewed seven trials comparing autologous vs. allogeneic transfusions; only twowereprospectiverandomizedtrialswitharound80patientsoneacharm.Thismeta- analysis suggested at least a twofold increase in postoperative infections in patients hav- ing allogeneic transfusions of 1–4 units. Sethna NF, Zurakowski D, Brustowicz RM, Bacsik J, et al. (2005) Tranexamic acid reduces intraoperative blood loss in pediatric patients undergoing scoliosis surgery. Anesthesiology 102:727 – 32 A recent and well done protocol that demonstratesa greater than 40% reduction in bleed- ing during spine surgery by using tranexamic acid. There was a clear trend to lower trans- fusion rates in the tranexamic group; however, it did not reach statistical significance. Tobias JD ( 2004) A review of intrathecal and epidural analgesia after spinal surgery in children. Anesthes Analg 98(4):956 – 65 A close look into the pediatric field of post spine surgery analgesia by an expert in pediat- ric orthopedic anesthesia. An interesting view of the use of regional anesthesia and spinal opioids. References 1. American Society of Anesthesia (2005) Guidelines on intraoperative monitoring. http:// www.asahq.org/publicationsandservices/standards/02.pdf. Web site accessed Jan 05, 2005 2. Anonymous (2003) Best Practice 7(2):1–6 3. Apfelbaum RI, Kriskovich MD, Haller JR (2000) On the incidence, cause, and prevention of recurrent laryngeal nerve palsies during anterior cervical spine surgery. Spine 25(22): 2906–12 4. Banoub M, Tetzlaff JE, Schubert A (2003) Pharmacologic and physiologic influences affect- ing sensory evoked potentials: implications for perioperative monitoring. Anesthesiology 99(3):716–37 5. Beaumont AC (2003) Blood transfusion: Reducing use, increasing safety. CPD Anesthesia 5(1):7–12 6. Ben-David B, Taylor PD, Haller GS (1987) Posterior spinal fusion complicated by posterior column injury. A case report of a false-negative wake-up test. Spine 12(3):540–3 7. Blake DW, Hogg MN, Hackman CH, et al. (1998) Induction of anaesthesia with sevoflurane, preprogrammed propofol infusion or combined sevoflurane/propofol for laryngeal mask insertion: cardiovascular, movement and EEG bispectral index responses. Anaesthesia Intensive Care 26(4):360 – 5 8. Blanloeil Y, Roze B, Rigal JC, Baron JF (2002) Hyperchloremic acidosis during plasma vol- ume replacement. Ann Francaises Anesth Reanim 21(3):211–20 9. Blumenthal S, Min K, Nadig M, Borgeat A (2005) Double epidural catheter with ropivacaine versus intravenous morphine: A comparison for postoperative analgesia after scoliosis cor- rection surgery. Anesthesiology 102:175–80 10. Colomina MJ, Bag´o J, Pellis´e F, et al. (2004) Preoperative erythropoietin in spine surgery. Eur Spine J 13(1):S40–S49 11. Cheng MA, Todorov A, Tempelhoff R, et al. (2001) The effect of prone positioning on intra- ocular pressure in anesthetized patients. Anesthesiology 95:1351–5 12. De La Cruz JP, Carmona JA, Paez MV, et al. (1997) Propofol inhibits in vitro platelet aggrega- tion in human whole blood. Anesth Analg 84:919–21 13. DiCindio S, Schwartz DM (2005) Anesthetic management for pediatric spinal fusion: Impli- cations of advances in spinal cord monitoring. Anesthesiol Clin N Am 23:765–87 14. Dimick JB, Lipsett PA, Kostuik JP (2000) Spine update: Antimicrobial prophylaxis in spine surgery. Basic principles and recent advances. Spine 25(19):2544–48 15. Dubos J, Mercier C (1993) Problemes anesthesiques et reanimation postoperatoire pour la chirurgie des scoliosis. Agressologie 34(1):27–32 412 Section Peri- and Postoperative Management . function of the spinal cord and nerve roots. Usually the spinal cord, brachial plexus roots, and L5 and S1 can be evaluated by asking the patient to move their hands and feet. The WUT is an outstanding. [26]. Recently, the use of aprotinin was associated with a doubling of the risk of renal failure, a 55% increased risk of myocardial infarction and a 181% increase in the risk of stroke in cardiac. active bleeding and: Each unit of FFP contains 2–4 mg of fibrinogen/ml; therefore each FFP unit delivers the equivalent of 2 units of cryoprecipitate PT or PTT 1.5 times that of control subjects;